Feeding behavior of the mussel Mytilus spp.: Responses to...

of 186/186
vaga 1 Feeding behavior of the mussel Mytilus spp.: Responses to the natural variability of seston and to toxic phytoplankton ingestion Conducta alimentària del musclo Mytilus spp.: Respostes a la variació natural del sèston i a la ingestió de fitoplàncton tòxic Eva Galimany Sanromà Barcelona, 2010
  • date post

    02-Dec-2019
  • Category

    Documents

  • view

    0
  • download

    0

Embed Size (px)

Transcript of Feeding behavior of the mussel Mytilus spp.: Responses to...

  • vaga 1

    Feeding behavior of the mussel Mytilus spp.: Responses to the natural variability of seston and

    to toxic phytoplankton ingestion

    Conducta alimentària del musclo Mytilus spp.: Respostes a la variació natural del sèston i

    a la ingestió de fitoplàncton tòxic

    Eva Galimany Sanromà

    Barcelona, 2010

  • 2

  • Front page:

    Master of Catherine of Clèves

    Border of St. Ambrose framed between mussels and a crab, c. 1440

    Miniature on parchment from “Book of Hours of Catherine of Clèves”

    Morgan Library and Museum, MS. 917, p. 224, New York

    Image obtained from the book:

    Dezallier d’Argenville, A.-J. (2009). Moluscos, Conchiglie, Conchas. Taschen, 216 pp.

    3

  • 4

  • Feeding behavior of the mussel Mytilus spp.: responses to the

    natural variability of seston and to toxic phytoplankton ingestion

    Conducta alimentària del musclo Mytilus spp.: respostes a la variació

    natural del sèston i a la ingestió de fitoplàncton tòxic

    Memòria presentada per

    EVA GALIMANY SANROMÀ

    per optar al grau de Doctor per la Universitat de Barcelona,

    realitzada al IRTA St. Carles de la Ràpita.

    Tesi inscrita al Departament de Biologia Cel·lular, Facultat de Biologia,

    Universitat de Barcelona

    Directora: Dra. Montserrat Ramón Herrero

    Tutora: Dra. Mercè Durfort Coll

    Programa de doctorat: Aqüicultura

    Bienni: 2004 - 2006

    Doctoranda Directora Tutora

    Eva Galimany Sanromà Dra. Montserrat Ramón Herrero Dra. Mercè Durfort Coll

    IRTA

    Centre Oceanogràfic de Balears Facultat de BiologiaSant Carles de la Ràpita

    Universitat de BarcelonaInstituto Español de Oceanografía

    Barcelona, 2010

    5

  • 6

  • ACKNOWLEDGEMENTS

    AGRAïMENTS

    I

  • Acknowledgements

    Agraïments

    I would like to thank,

    · Montse, my PhD director, per la seva constant ajuda en tot i per tot, els seus bons

    consells, i que sempre hagi cregut en mi. No podria haver tingut millor directora de tesi, I’ll be

    always grateful,

    · all those aussies, specially Tim, who, with blind acceptance, introduced me to marine

    sciences,

    · Irrintzi, Udane y su equipo de personal filtrador, por abrirme las puertas del mágico

    mundo de la filtración de los bivalvos,

    · Richard and the MBA, for sharing his wide knowledge and helping me every time that

    I’ve needed,

    · Gary, Hélène and the Milford Laboratory staff, for teaching me to trust in people and

    making of science and stats an enjoyable party,

    · Inke, for her wise advises, perseverance and bravery in a world still not made for

    scientific women,

    · the patience of all the people I’ve talked about above when I’ve broken lab material or

    messed up the labeling of samples,

    · ICM-CSIC, for making me feel at home,

    · Dra. Mercè Durfort, per donar-me sempre el seu suport i compartir els seus

    coneixements histopatològics,

    · la meva família, en especial als meus pares, qui m’han ajudat sempre, en qualsevol

    lloc i qualsevol moment,

    · last but not least, George, who has always believed in my scientific career and has

    helped and backed me up during all these years; and Iris, for reminding me, every single day,

    that science is part of our lives. We all do science, everywhere, at any time, spontaneously,

    since early stages of life.

    II

  • The Sea Shell

    BY MARIN SORESCU

    I have hidden inside a sea shell

    but forgotten in which.

    Now daily I dive,

    filtering the sea through my fingers,

    to find myself.

    Sometimes I think

    a giant fish has swallowed me.

    Looking for it everywhere I want to make sure

    it will get me completely.

    The sea-bed attracts me, and

    I’m repelled by millions

    of sea shells that all look alike.

    Help, I am one of them.

    If only I knew, which.

    How often I’ve gone straight up

    to one of them, saying: That’s me.

    Only, when I prised it open

    it was empty.

    TRANSLATED BY MICHAEL HAMBURGER

    III

  • To George, my parents,

    and scientific women, specially scientific moms.

    IV

  • INDEX

    V

  • Index

    Introduction 1

    1. The importance of bivalves 2

    1.1 Mussels as keystone species 2

    1.2 Mussels as sentinel species and quality programs 3

    2. Basic knowledge on the genus Mytilus 3

    2.1 Anatomy 3

    2.2 Geographic distribution 5

    3. Feeding 8

    3.1 Particle selection 8

    3.2 Harmful Algal Blooms (HAB) 12

    4. The immune system 13

    5. Mussel culture in the Ebro Delta 15

    Objectives 18

    Report of the Director 20

    Results 24

    Chapter I. Feeding behavior of the mussel Mytilus galloprovincialis (L.) in a

    Mediterranean estuary: a field study 25

    Chapter II. Pre-ingestive selection of mussels, Mytilus galloprovincialis (L.), grazing on

    natural phytoplankton in a N.W. Mediterranean estuary 43

    Chapter III. First evidence of fiberglass ingestion by a marine invertebrate (Mytilus

    galloprovincialis L.) in a N.W. Mediterranean estuary 60

    Chapter IV. The effects of feeding Karlodinium veneficum (PLY # 103; Gymnodinium

    veneficum Ballantine) to the blue mussel Mytilus edulis 66

    Chapter V. Pathology and immune response of the blue mussel (Mytilus edulis L.) after

    an exposure to the harmful dinoflagellate Prorocentrum minimum 75

    Chapter VI. Experimental exposure of the blue mussel (Mytilus edulis, L.) to the toxic

    dinoflagellate Alexandrium fundyense: histopathology, immune responses, and recovery 85

    Discussion 96

    1. The ecological role of mussels 97

    2. Validation of methods 98

    VI

  • VII

    2.1 The in situ feeding experiments 99

    2.2 The HAB experiments 102

    3. Benefits of research 104

    General Conclusions 106

    Resum de la tesi (summary in catalan) 109

    1. Introducció 110

    2. Objectius 119

    3. Resultats 120

    4. Discussió 126

    5. Conclusions 133

    Bibliography 135

  • INTRODUCTION

    1

  • Introduction

    INTRODUCTION

    1. The importance of bivalves Bivalves are marine and freshwater molluscs which are distributed around the world.

    The ecological role of bivalves has been widely discussed and they can provide different

    ecosystem functions. For example, bivalves can improve the water quality of the ecosystems

    due to their filter-feeding capacity to mitigate the effects of eutrophication (Ostroumov, 2005;

    Coen et al., 2007; Lindahl and Kollberg, 2009). They can also contribute to the creation of

    habitat heterogeneity for fishes and invertebrates (Gutiérrez et al., 2003; Coen et al., 2007), and

    to the transfer of chemical elements and nutrients from the water and seston to other trophic

    levels and the sediments, linking benthic and pelagic ecosystems (Prins et al., 1998; Newell,

    2004). The ecological impact of some bivalves in shallow, coastal embayments may not be

    solely as regulators of primary production but also secondary production (Lonsdale et al., 2009).

    Bivalves can also contribute to the stabilization of benthic or intertidal habitats, i.e. mussel beds

    (Coen et al., 2007; Dumbauld et al., 2009).

    Some bivalve species also have economic roles through fisheries, aquaculture, and the

    ornament industry. Of all bivalves cultivated around the world, mussels are ideal for aquaculture

    due to their wide distribution and adaptability. Mussels are harvested for human consumption

    worldwide (Figueiras et al., 2002; Mortensen et al., 2006) with production of the species

    Mytilus galloprovincialis and M. edulis above 300.000 tons per year (FAO, 2009), mostly

    cultured in Spain, the leading mussel producer in Europe (Pérez Camacho et al., 1991; Keldany,

    2002).

    1.1 Mussels as keystone species

    Mussels feed on several types of particles suspended in the water column but

    phytoplankton has traditionally been considered to be their main source of food (Mason, 1971).

    As a consequence of their filter-feeding activity, mussels can change the phytoplankton

    community; the filtration of bivalves has been shown to control phytoplankton growth, in what

    is often referred to as “top-down” processes. When not sufficiently grazed, phytoplankton

    populations can bloom excessively, leading to the deterioration of water quality and

    eutrophication. But, mussels also have the potential to promote primary production by nutrient

    release in the biodeposits, establishing a balance between phytoplankton grazing and release of

    nutrients (Asmus and Asmus, 1991). This event is especially relevant in shallow-water

    ecosystems, where the ratio of benthic surface area to water volume is high (Dame, 1996). The

    relationship established between both trophic levels is known as benthic-pelagic coupling;

    benthic bivalves transport particulate matter from the water column to the benthos transferring

    2

  • Introduction

    3

    organic nitrogen and carbon from the water column to the sediments (Verwey, 1952; Doering et

    al., 1986; Dame et al., 1989). Due to all these interactions, mussels are considered keystone

    species because they have important ecological roles maintaining water quality and participating

    in the cycling of nutrients from the water column to the benthos (Prins and Smaal, 1994;

    Ostroumov, 2005).

    1.2 Mussels as sentinel species and quality programs

    As a consequence of the ability of bivalves to accumulate anthropogenic pollutants and

    increasing industrial effluents in the estuarine and coastal environments, Goldberg (1975)

    proposed a “Mussel Watch” monitoring program to assess spatial and temporal chemical

    contamination in North America, using mussels as sentinel organisms. After such proposal, an

    international “Mussel Watch” program was developed in order to assess the levels of certain

    contaminants in bivalves collected from coastal marine waters throughout the world

    (International Musselwatch Committe, 1992). Sentinel monitoring species should comply with

    several criteria (Morgado and Bebbiano, 2005), such as being abundant, accessible, sedentary

    and cosmopolitan, large enough to provide sufficient tissue for biological analysis, easy to

    identify and available all year long. In addition, they should respond to contamination but be

    robust enough to be present in polluted environments and be easy to maintain in the laboratory

    for laboratory studies. Mussels are one of the few species that satisfy all these characteristics. In

    detail, the genus Mytilus tolerates polluted waters, wide variations in salinity and temperature,

    and efficiently accumulates various trace metals (Rainbow, 1995; Morgado and Bebbiano,

    2005).

    When consumed, bivalves can have negative impacts on human health. Due to filter-

    feeding, bivalves can be vectors of diseases related to the accumulation of bacteria, viruses,

    pesticides, biotoxins, industrial wastes, toxic metals, and petroleum derivatives from the water

    column, with subsequent public health concerns. Thereafter, water quality programs of

    shellfishing areas were also established in many countries around the world in order to protect

    public health (Gosling, 1992).

    2. Basic knowledge on the genus Mytilus

    2.1 Anatomy

    Mussels of the genus Mytilus are sessile bivalve mollusks which live in sea water and

    estuarine habitats. They posses two shell valves with similar size and shape which protects them

    from predators and the environment. Figure 1 shows the gross anatomy of a mussel.

  • Introduction

    1

    2 3

    Fig. 1: Basic anatomy of Mytilus edulis. 1.1. Photograph of the soft part. 1.2. Illustration of the anatomy. 1.3. Illustration of a cross-section; bold parallel lines in fig. 1.1

    and 1.2 show location of the cross-section. From Howard et al. (2004), illustrations by A. J. Lippson, Bozman, M.D.

    4

  • Introduction

    The viscera of the mussels are attached to the shells by adductor muscles, the mantle

    edge and small points of attachment every few millimeters between the mantle and the inner

    shell (Bayne et al., 1979). The mantle consists of two lobes of tissue which enclose the bivalve

    within the shell. Between the mantle and the internal organs is the mantle cavity. In mussels, the

    mantle is the main site for the storage of nutrient reserves but it also contains most of the gonad.

    When ripe, the mantle is thick, but after the release of the gametes, it gets thin and transparent

    (Gosling, 2003). The gills (or ctenidia) are flat filibranch structures that are suspended from the

    dorsal margin of the mantle with two different physiological functions: respiration and feeding

    (Bayne et al., 1976; Gosling, 1992). The foot of the mussel is a long and highly mobile

    muscular structure which larvae and adults use to detect a suitable substrate and attach to it.

    Moreover, by means of the byssal gland, located in the foot, mussels secrete byssal threads for

    attachment to the substrate.

    The alimentary canal consists of mouth, esophagus, stomach, intestine, digestive tubules

    and anus. The excretory organs of mussels are the pericardial glands and a pair of kidneys. The

    heart, in the mid-dorsal region of the body, pumps the hemolymph around the body. The

    circulatory system of mussels is open, which means, that the hemolymph in the sinuses bathe

    the tissues directly (Gosling, 2003).

    2.2 Geographic distribution

    The genus Mytilus includes the species: M. edulis (Linnaeus 1758), M. trossulus (Gould

    1850), M. galloprovincialis (Lamark 1819), M. chilensis (Hupé 1854), M. coruscus (Gould

    1861), M. californianus (White 1937), M. platensis (o’Orbigny 1846), M. planulatus (Lamarck

    1819), and M. desolationis (Lamy 1936) (Gosling, 1992). Prior to the use of electrophoresis,

    taxonomy was based on shell morphology, a characteristic that varies with numerous

    environmental factors. Consequently, there is confusion in the literature concerning occurrence

    of the different species. Nowadays, molecular techniques allow more accuracy in mapping their

    distribution around the world. Figure 2 shows the worldwide distribution of four species of

    Mytilus. In detail, the species M. galloprovincialis and M. edulis occur globally, whereas M.

    trossulus and M. californianus are restricted to the northern hemisphere. Recent studies clarified

    the occurrence of M. galloprovincialis in the Galician Rías (N. W. Iberian Peninsula), which

    was thought to be M. edulis (Crespo et al., 1990; Sanjuan et al., 1990). Thereafter, along the

    European coast, the species M. galloprovincialis inhabits warmer waters such as the

    Mediterranean Sea and the coasts of Spain, France and southern Britain whereas M. edulis is

    found in higher latitudes. Both Mytilus species overlap and hybridize naturally in the European

    Atlantic coast (Beaumont et al., 2004).

    5

  • Introduction

    Fig. 2: Global distribution of four species of the genus Mytilus. H means areas of hybridization. From

    Gosling (1992).

    Mussels of the genus Mytilus inhabit the low and intertidal zone of temperate seas

    globally. Of all the species of the genus, the blue mussel M. edulis has the widest distribution,

    from mild subtropical to Arctic regions, from high intertidal to subtidal regions, from estuarine

    to true marine conditions, and from sheltered to extremely wave-exposed shores (Gosling,

    2003). The distribution and abundance of mussels are affected by the following physical and

    biological factors:

    1. Physical factors: bivalve mollusks are poikilothermic animals, which mean that they

    produce small amounts of heat during metabolism and are thermally dependant on the

    environment (Dame, 1996). Thereafter, temperature might be considered as the most important

    physical factor that regulates the distribution of mussels; it sets limits on the spatial distribution

    of bivalves and affects every aspect of their biology. The different Mytilus species tolerate

    different temperature thresholds. M. galloprovincialis cannot survive sea water temperatures of

    or beyond 26ºC over extended periods of time (Anestis et al., 2007) and inhabits warmer waters

    than M. edulis (Beaumont et al., 2004). Similarly, M. trossulus is more thermally sensitive than

    M. galloprovincialis and heat shock proteins, which expresion increases when cells are exposed

    to high temperatures, are synthesized at 23ºC, a lower temperature than for M. galloprovincialis

    (Hofmann and Somero, 1996). Low temperature tolerances also vary among Mytilus species;

    whereas M. edulis survive winter temperatures lower than -10ºC, M. californianus cannot

    tolerate freezing conditions (Williams, 1970; Seed and Suchanek, 1992). Endogenous cellular

    6

  • Introduction

    estress proteins in M. edulis vary seasonally and correlate positively with seasonal changes in

    both environmental temperature and thermal tolerance (Chapple et al., 1998). Water temperature

    is known to directly affect the physiological status of bivalves regulating feeding behavior,

    growth and spawning of mussels (Denis et al., 1999; Suárez et al., 2005). As an example, when

    temperature exceeds 25 ºC, filtration rates fall significantly in M. galloprovincialis and in M.

    edulis (Gonzalez and Yevich, 1976; Anestis et al., 2007).

    Salinity is also a very important limiting factor in coastal and estuarine bivalves.

    Nevertheless, Mytilus spp. are considered euryhaline, and they are able to adapt to a wide range

    of salinities. In particular, a range between 4 and 40 ‰ has been reported for M. edulis (Bayne

    et al., 1976).

    Other factors may be considered as limitations for bivalve distribution such as depth,

    tides, currents, type of substrate, turbidity, etc. (Dame, 1996; Gosling, 2003).

    2. Biological factors: food and predators, together with pathological conditions and

    parasites, are probably the most important source of natural mortality in bivalves. They have the

    potential to affect population size structure in addition to overall abundance and local

    distribution patterns (Seed and Suchanek, 1992).

    Of all the particles suspended in the environmental water, mussels mainly feed on

    phytoplankton although they can also feed on bacteria, zooplankton, detritus as well as

    dissolved organic material (Mason, 1971; Gosling, 2003). Potential predators of adult mussels

    include gastropods, starfish, sea urchins, crabs, fish, birds and humans (Gosling, 2003; Dame,

    1996; Kamermans et al., 2009). Predation of bivalve larvae is mostly due to filter-feeding

    invertebrates, including bivalves (Dame, 1996; Gosling, 2003). There must be equilibrium for

    both the food source and the predators to allow mussels to survive, grow and reproduce in each

    inhabiting area.

    Most of the pathological conditions of marine bivalves are caused by different types of

    infectious agents and parasites such as viruses, bacteria, fungi, trematodes, annelids, copepods

    and different protists (Lauckner, 1983). Some of the most common pathological conditions

    found in mussels include marteiliosis caused by the parasite Marteilia refringens, the lesions

    caused by the copepod Mytilicola intestinalis and the trematode Proctoeces maculatus, and

    disseminated neoplasia (Lauckner, 1983; Kim and Powell, 2007).

    In mussels, the shell provides an excellent substrate for the settlement of fouling

    organisms, which can cause significant mortality due to dislodgement, as a consequence of the

    increase in weight and clogging. In addition, some of the epibionts are filter-feeders competing

    with mussels for food. Nevertheless, under conditions where food is not a limiting factor,

    interespecific competion by the epibionts should not significantly limit the yield of mussels

    7

  • Introduction

    (Lesser et al., 1992). Almost 100 invertebrate species have been identified as fouling organisms

    on suspended mussel rope cultures (Perera et al., 1990; Hickman, 1992).

    3. Feeding

    3.1 Particle selection

    Mussels are sessile filter-feeding bivalves which depend on the water where they are

    submerged for food supply. The water is forced through an inhalant siphon to get into the

    mussel (Fig. 3). Then, lateral cilia on the gills maintain a flow of water through the mantle

    cavity and gills to allow the filtration of the water. The cilia remove the large particles and trap

    them in a fine mucus layer, transporting them towards the labial palps, two triangular structures

    located on each site of the mouth. The smaller particles can either escape or are captured by the

    cilia. The filtered water exits the mussel through the exhalant siphon (Riisgård et al., 1996;

    Gosling, 2003; Ward and Shumway, 2004). The palps direct the particles retained by the gills

    towards the mouth, which leads into the stomach by means of a narrow esophagus.

    Fig. 3: Principal pathways of current flow and particle transport in Mytilus edulis. Water enters the

    bivalve via the inhlant siphon (IS). The frontal surface of the gill (G) is exposed to a postero-anterior flow

    at the ventral margin; the rest of the frontal surface is swept by a ventro-dorsal flow, with a progressive

    through component to the abfrontal region. Solid arrows indicate current flow prior to passage across

    gills. Water exits the gill through the exhalant siphon (ES) (open arrows). Arrowheads represent particle

    transport to the ventral groove of the gills (VG); (P labial palp). From Beninger and St.-Jean (1997a).

    8

  • Introduction

    Extracellular digestion takes place in the stomach by enzymes released from the rotating

    crystalline style (Fig. 4). Intracellular digestion occurs immediately after the stomach in the

    digestive gland. The unabsorbed waste material from the digestive gland is formed into fecal

    pellets which are directed through the anus, towards the exhalant siphon and released (Gosling,

    2003).

    Fig. 4: Part of the digestive system of a bivalve showing rotation of the crystalline style. From Gosling

    (2003).

    Gills act like sieves that can retain different types of particles within the size threshold

    of 1 to 6000 µm (Lehane and Davenport, 2006) although the retention efficiency differs

    according to particle size. Mytilus edulis can retain 1-2 µm particles with an efficiency of 50%

    (Jørgensen, 1975), 2 µm particles with an efficiency of 75-90%, and particles larger than 6 µm

    are retained with 100% efficiency (Møhlenberg and Riisgård, 1978).

    However, bivalves do not ingest everything that is retained by the gills (Gosling, 2003).

    Pre-ingestive selection may occur on the gills and labial palps and results in the production of

    pseudofeces (Kiørboe and Møhlenberg, 1981; Shumway et al., 1985; Ward et al., 1998; Zemlys

    et al., 2003) (Fig. 5). The selection of particles can also occur in the stomach by retaining some

    particles longer to increase time for extracellular digestion or by directing some particles to the

    digestive gland (Bricelj et al., 1984; Shumway et al., 1985; Brillant and MacDonald, 2003;

    Ward and Shumway, 2004). The stomach can also sort and direct ingested particles directly to

    the intestine based on large sizes, low nutrition, high density or certain chemical properties

    9

  • Introduction

    (Morton, 1973; Ward and Shumway, 2004). These processes are known as post-ingestive

    selection (Fig. 5).

    water particles

    ingested particles

    stomach(extracellular

    digestion)

    gills

    feces

    rejected particles

    pseudofeces

    intestine

    anus

    digestive gland

    (intracellulardigestion)

    waste material

    pre-ingestiveselection

    post-ingestiveselection

    Fig. 5: Diagram of particle selection by mussels.

    Food that is finally directed from the stomach towards the digestive gland (including

    digestive ducts and tubules) goes under intracellular digestion and the waste material is

    redirected to the stomach (Fig. 6).

    Food that passes directly from the stomach through the intestine results in poorly

    digested material (Widdows et al., 1979a). Feces composed of the waste material from the

    digestive gland are named glandular feces; intestinal feces are produced when seston

    concentrations exceed the maximum digestible concentration and are rejected after deficient

    digestion in the intestine (Widdows et al., 1979a). The selection of particles by bivalves is a

    strategy to maximize the quality of the diet and to optimize the energy gain (Fig. 6).

    The selection criteria of filter-feeding bivalves can be affected by different

    characteristics of the available food particles. Inorganic particles, such as silt or glass, silica or

    alumina spheres are selected and rejected as pseudofeces (Kiørboe et al., 1980; Ward and

    Targett, 1989; Bayne et al., 1993). Moreover, the production of pseudofeces increases along

    10

  • Introduction

    with the inorganic content of the water (Kiørboe et al., 1980), or when the ingestive capacity of

    bivalves is overloaded (Beninger and St-Jean, 1997b).

    Fig. 6: Section of the digestive gland showing intracellular digestion of material coming from the stomach

    (solid arrows) and redirection of waste material to stomach (broken arrows). From Gosling (2003).

    Pre-ingestive selection has also been reported in several other occasions. When feeding

    on different species of phytoplankton, diatoms are preferentially rejected over dinoflagellates

    and flagellates (Shumway et al., 1985; Bougrier et al., 1997). Algal size, shape, flexibility and

    quality can also affect the capture efficiency of particles (Bougrier et al., 1997; Defossez and

    Hawkins, 1997; Ward et al., 2003). As a consequence, the filtration activity of a mussel

    population and its selective efficiency on various taxa interacts with the phytoplankton

    community of the ecosystem causing changes in the species and size distributions (Asmus and

    Asmus, 1991; Noren et al., 1999; Dolmer, 2000; Petersen et al., 2008). Therefore, mussels have

    the potential to exert significant top-down control of phytoplankton (Strohmeier et al., 2008;

    Trottet et al., 2008).

    High density mussel populations, such as farmed mussels, grazing on phytoplankton can

    have a top-down control effect on the communities of phytoplankton. Nevertheless, most of the

    ingested organic matter is rapidly recycled to the water column as inorganic nutrients, which

    would be expected to stimulate phytoplankton growth. As a consequence, the net effect of

    mussel farming on phytoplankton dynamics may increase phytoplankton turnover and overall

    production (Nizzoli et al., 2005).

    11

  • Introduction

    3.2 Harmful Algal Blooms (HAB)

    Among the many different phytoplankton species that mussels feed on there are some

    that are considered toxic. Toxic microalgal proliferations in aquatic ecosystems, referred to as

    Harmful Algal Blooms (HABs), were first recorded by ancient civilizations and have been

    observed ever since (Fogg, 2002; Landsberg, 2002). HABs appear to be increasing in

    geographic distribution and intensity, along with their effects and consequences on the

    ecosystems and their organisms (Landsberg, 2002). As an anecdote, the movie director Alfred

    Hitchcock witnessed an episode of HAB and how shearwaters, oceanic birds that feed on

    anchovies, flew inland toward the city of Capitola, California (USA), dying by the dozens as

    they crashed into buildings and cars. The anchovies had eaten toxic algae, which produced

    domoic acid, a neurotoxin. The birds in turn had eaten the fish, affecting the birds' ability to fly.

    This incident was used as research material for the classic thriller about a coastal town

    terrorized by deranged birds, The Birds (1963, based on the 1952 novella Daphne du Maurier).

    Different taxa of phytoplankton, including dinoflagellates, diatoms, raphidophytes,

    prymnesiophytes, silicoflagellates, ciliates, and cyanobacteria have the potential to be toxic to a

    wide variety of organisms, including bivalves (Shumway, 1990; Fogg, 2002; Landsberg, 2002).

    The filter-feeding activity of bivalves makes them susceptible to ingest toxic algae whenever

    present in the environmental water. It is known that mussels can use toxic dinoflagellates as a

    food source in the absence of other phytoplankton cells (Bricelj et al., 1993). Moreover, when a

    toxic algal species proliferates, it becomes predominant among the entire phytoplankton

    community of the ecosystem, becoming almost the only available food source for bivalves.

    As bivalves are grown under natural conditions the biological quality of the growing

    water is very important. Bivalves can be affected by different types of algal toxins (Shumway,

    1990; Landsberg, 2002) and their responses to the different harmful algae seem to be species-

    specific (Shumway et al., 2003; Hégaret et al., 2007a). In detail, several biological effects of

    toxic microalgae upon mussels have been described (Table 1). Moreover, toxins produced by

    HAB species can accumulate in bivalve tissues and affect their predators, passing the toxins to

    higher trophic levels, including top predators in food webs such as humans (Azanza and Taylor,

    2001; Landsberg, 2002; Jester et al., 2009).

    Of all bivalves consumed by humans, mussels accumulate toxins more rapidly than any

    other species (Shumway et al., 1990; Gosling, 2003). Therefore, to protect public health,

    monitoring and management programs for bivalve toxins have been implemented (Shumway et

    al., 1990; Rehnstam-Holm and Hernroth, 2005). HAB-related closures of shellfish harvesting

    can result in great economic losses for the aquaculture industry (Shumway, 1990; EUROHAB,

    1998).

    12

  • Introduction

    Effects Microalgal specie Bibliography

    Decrease of clearance rate

    Alexandrium monilatum Alexandrium tamarense Gymnodinium aureolumKarenia brevis Prorocentrum lima

    Pate et al. (2005) Lesser and Shumway (1993) Widdows et al. (1979b) Leverone et al. (2007) Pillet and Houvenhagel (1995)

    Reproductive failure Aureococcus anophagefferens Chrysochromulina polylepis

    Bricelj and Kuenstner (1989) Granmo et al. (1988)

    Inhibition of byssus production

    Alexandrium tamarense Heterocapsa circularisquama

    Shumway et al. (1987) Matsuyama et al. (1998)

    Growth suppression Aureococcus anophagefferens Chrysochromulina polylepis

    Bricelj et al. (2001) Nielsen and Strømgren (1991)

    Shell valve closure Alexandrium tamarense Heterocapsa circularisquama

    Shumway and Cucci (1987) Matsuyama et al. (1998)

    Mortality

    Aureococcus anophagefferens Gonyalaux spp. Gymnodinium aureolum Gyrodinium corsicum Rhizosolenia chunii

    Bricelj et al. (2001) O’Sullivan (1978) Cross and Southgate (1980) ICES (1999) Parry et al. (1989)

    Table 1: Effects of different toxic microalgal species upon mussels (Mytilus spp.).

    Recent research has reported viable cells and temporary cysts from different harmful

    algae in the feces of several bivalves. Thus, in addition to the noxious effects of HABs to

    bivalves and predators, there is a potential transport of harmful algae via relocation of bivalves

    (Hégaret et al., 2008).

    4. The immune system

    Circulating hemolymph cells, known as hemocytes, constitute the major medium for

    immune defense of bivalves. These ubiquitous cells are found in the open circulatory system of

    bivalves including heart, hemolymph vessels and variably sized sinuses localized in the major

    organs (Auffret, 2005). The capability of hemocytes to discriminate between self and non-self

    is, in part, based on the presence of lectins on the surface of the cells (Renwrantz, 1990).

    Nevertheless, it is suggested that the type of reaction depends upon the nature of the particles

    presented, i.e.: hemocytes exhibit chemotactic and chemokinetic reactions when exposed to

    bacterial products (Schneeweiß and Renwrantz, 1993).

    Hemocytes can be divided into two main groups: basophilic and eosinophilic cells; and

    hemocytes may be either agranular or granular (Fig. 7).

    13

  • Introduction

    2

    31

    Fig. 7: 7.1. Light microscopy image of agranular (A) and granular (G) hemocytes of Mytilus edulis. Scale

    bar = 20 µm. From Rasmussen et al. (1985). 7.2 and 7.3. Electron microscopy images of two basophilic

    and an eosinohilic hemocyte respectively of M. edulis. Scale bar = 1 µm. From Pipe et al. (1997).

    Basophilic hemocytes can be divided into small (4-5 µm) and large (7-9 µm) cells

    (Moore and Lowe, 1977; Bayne et al., 1979; Pipe, 1990a). The small basophilic hemocytes are

    generally spherical with a spherical nucleus and hyaline cytoplasm. The larger basophilic cells

    assume an irregular appearance with granules and vacuoles in the cytoplasm, which

    considerably vary in diameter. The eosinophilic hemocytes, or granulocytes, are often spherical

    and filled with spherical granules (0.5-1.0 µm) which range from neutrophilic to acidophilic in

    staining properties (Moore and Lowe, 1977; Pipe, 1990a). Table 2 shows a brief simple

    classification of the hemocytes with their basic characteristics.

    Once hemocytes have recognized a non-self particle, they can either phagocytose the

    particles or release a range of antimicrobial molecules including reactive oxygen metabolites

    (Pipe, 1992; Dyrynda et al., 1998; Wootton et al., 2003a). The granules of the different types of

    hemocytes posses a wide range of hydrolytic enzymes, such as phosphatases, esterases,

    peroxidases, proteinases, glycosidases and sulphatases, showing that these granules are a type of

    lysosomes (Moore and Lowe, 1977; Pipe, 1990b; Carballal et al., 1997a). Nevertheless, mussel

    hemocytes are functionally heterogeneous as the eosinophilic cells posses the ability for

    phagocytosis whereas the basophilic hemocytes seem to be non or much less phagocytic (Pipe,

    1990b; Carballal et al., 1997b; Pipe et al., 1997). The release of lysosomal enzymes by granular

    hemocytes is accompanied by degranulation of the cell (Foley and Cheng, 1977).

    14

  • Introduction

    Type of hemocyte Basophilic Eosinophilic (Granular)

    small large

    Shape spherical irregular spherical

    Granules no granules different diameters spherical (0.5-1.0 µm)

    Table 2: Classification and characteristics of the different types of hemocytes.

    The immune system of invertebrates is considered non-specific (or innate), which

    means, that the immune system is wide-ranging providing immediate defense recognizing and

    responding to noxious particles and environments in a generic way. In contrast to the adaptive

    immune system, it does not confer long-lasting or protective immunity to the host. Nevertheless,

    bivalves possess adaptive strategies against non-favorable ecosystems. In accordance, when

    comparing the immune function of the mussel M. edulis, the edible cockle Cerastoderma edule

    and the razor-shell Ensis siliqua, mussels showed a higher level of immunological vigor

    probably linked to their considerable resilience to adverse environmental conditions (Wootton et

    al., 2003b). Mussels inhabiting contaminated areas are more susceptible to infections because of

    the immunosuppression caused by pollution stress (Pipe and Coles, 1995; Dyrynda et al., 1998).

    Nevertheless, Mytilus populations have been reported to survive exposure to heavy oil pollution

    in contrast with other bivalves (Dyrynda et al., 2000; Ordás et al., 2007). In addition, it has been

    reported that bivalves periodically exposed to HABs become more resistant to the toxic effects

    of the different microalgal toxins to which they are commonly exposed (Bricelj et al., 2005).

    5. Mussel culture in the Ebro Delta

    Mussel (Mytilus galloprovincialis) culture along the Spanish Mediterranean coast is

    traditional in the regions of Catalonia, Valencia and Menorca, and is now also developing in

    Andalusia (Ramón et al., 2005) (Fig. 8). Among these regions, the major production occurs in

    Catalonia, specifically in the Ebro Delta bays (Ramón et al., 2005), with an average annual

    mussel production of about 3000 Tn (Fig. 9). At this site, mussel culture was initiated at the

    beginning of the 60’s and has been traditionally done ever since. Mussels are cultured in

    suspension from the 166 fixed rafts that are spread between the bays of Alfacs (90 farms) and

    Fangar (76 farms). The mussel rafts are rectangular wooden frames measuring 200 m x 15 m,

    from which 2-3 m long mussel ropes are suspended. They are anchored to the seabed by

    concrete girders, and only protrude from the surface of the water by 1 m.

    15

  • Introduction

    Málaga Almería

    Murcia

    Alicante

    Valencia

    Castellón

    Tarragona

    Barcelona

    1

    2

    4

    3

    Fig. 8: Map of the Spanish Mediterranean coast. Arrows and images of mussels show culture locations: 1.

    Ebro Delta; 2. Valencia; 3. Menorca; 4. Andalusia.

    The culture cycle begins during December and January by means of hanging collector

    ropes to obtain natural spat. The seed mussel reach 3 cm long some months later, at around

    October, when they are moved to the final ropes and remain there until reaching 6 cm in length

    during May - June.

    Despite the relatively low mussel production compared with Galicia and other world

    aquaculture centers, the Ebro Delta represents an important income for many families in the

    area. Moreover, the bays are embedded in a naturally protected area, which is the second largest

    wetland area in the western Mediterranean. This peculiarity makes it difficult to manage the

    bivalve cultivation given the protected area regulations in a widely variable environment such as

    this one.

    16

  • Introduction

    17

    0500

    1000150020002500300035004000

    1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

    Tn

    Year

    Fig. 9: Mussel production in the Ebro Delta bays (DAR, 2009).

  • OBJECTIVES

    18

  • Objectives

    OBJECTIVES

    The purpose of the present thesis was to study the feeding behavior of mussels in terms

    of selection and physiological components of absorptive balance, and to find out whether

    ingestion of toxic algae species modulates the immune function and causes pathological

    changes.

    To address this problem statement, two different mussel species, Mytilus

    galloprovincialis and M. edulis were used. Two different experimental setups were applied; the

    feeding experiments were performed in situ on top of a mussel raft under natural conditions, and

    mussels were experimentally exposed to toxic algae in aquaria under controlled laboratory

    conditions.

    Following hypotheses were formulated:

    Hypothesis I: The feeding behavior of mussels in Alfacs Bay (N.W. Mediterranean

    Sea) depends upon the quantity and quality of seston, which may vary in time.

    Hypothesis II: Mussels in Alfacs Bay (N.W. Mediterraean Sea) preferentially select

    some of the available suspended matter but not other.

    Hypothesis III: Harmful Algal Blooms modulate the immune function of mussels and

    cause pathological changes.

    This thesis is based on six chapters, which are referred to in the text by their Roman

    numeration.

    The hypotheses were tested by the following aims:

    1. Determine the main physiological parameters related to mussel feeding behavior by

    means of the biodeposition method in Alfacs Bay (Chapter I).

    2. Study the phytoplankton composition, determine the dominant taxa, and detect

    phytoplankton peaks during four periods of a culture cycle (from October to July) in a mussel

    aquaculture site (Alfacs Bay) (Chapter II and III).

    3. Study the selection and ingestion of available particles by mussels, including some

    toxic phytoplankton species, during four periods of a culture cycle in Alfacs Bay (Chapter II

    and III).

    4. Study the immunological and histological effects of three different toxic algae:

    Karlodinium veneficum, Prorocentrum minimum and Alexandrium fundyense on mussels using

    experimental laboratory exposures (Chapter IV, V and VI).

    5. Test possible recovery of mussels after experimental exposure to toxic Alexandrium

    fundyense (Chapter VI).

    19

  • REPORT OF THE DIRECTOR

    20

  • Report of the director

    REPORT OF THE DIRECTOR

    La Dra. Montserrat Ramón Herrero, directora de la tesi titulada: “Feeding behavior of

    the mussel Mytilus spp.: responses to the natural variability of seston and to toxic

    phytoplankton ingestion” (“Conducta alimentària del musclo Mytilus spp.: respostes a la

    variació natural del sèston i a la ingestió de fitoplàncton tòxic”) realitzada per Eva

    Galimany Sanromà,

    Informa de la implicació de la doctoranda en cada article científic desenvolupat per la

    present tesi i que cap dels articles, ni de les dades aquí presentades, han estat usades per la

    memòria d’una altra tesi.

    Chapter I. Article: “Feeding behavior of the mussel Mytilus galloprovincialis (L.) in

    a Mediterranean estuary: a field study”

    no publicat, la versió que es presenta en la memòria de tesi es la versió consensuada

    pels coautors i ja enviada a avaluar en una revista especialitzada internacional. El disseny

    experimental es va fer conjuntament entre la directora i la doctoranda. La doctoranda va

    intervenir en la construcció i posta a punt dels acuaris del sistema experimental i en el mostreig.

    L’asessorament del mètode de la biodeposició, usat pel càlcul dels components biològics del

    balanç d’absorció en bivalves correspon al Dr. Irrintzi Ibarrola, del Dpto. de Genética,

    Antropología Física y Fisiología Animal de la Facultad de Ciencia y Tecnología de la

    Universidad del País Vasco. La doctoranda va realitzar la cuantificació del contingut en sèston

    de l’aigua, i de les femtes i pseudofemtes del musclo, va treballar les dades, interpretar els

    resultats i redactar l’article, sota l’assessorament dels coautors (M. Ramón i I. Ibarrola).

    Chapter II. Article: “Pre-ingestive selection of mussels, Mytilus galloprovincialis

    (L.), grazing on natural phytoplankton in a N.W. Mediterranean estuary”

    no publicat, la versió que es presenta en la memòria de tesi es la versió revisada i

    consensuada pels diversos coautors i enviada a una revista especialitzada internacional. El

    disseny experimental es va fer conjuntament entre la directora i la doctoranda. La identificació

    d’espècies de fitoplancton es va dur a terme pel Dr. Delgado. La doctoranda va intervenir en la

    posta a punt del sistema experimental i el mostreig, la interpretació dels resultats i la redacció de

    l’article, assessorada pels coautors (M.Ramón y M. Delgado).

    21

  • Report of the director

    Chapter III. Article: “First evidence of fiberglass ingestion by a marine

    invertebrate (Mytilus galloprovincialis L.) in a N.W. Mediterranean estuary”

    publicat a la revista Marine Pollution Bulletin, l’índex d’impacte de l’any 2008 fou de

    2.562. El disseny experimental es va fer conjuntament entre la directora i la doctoranda. El Sr.

    José Manuel Fortuño, tècnic del Microscopi Electrònic de Rastreig (SEM), va realitzar el

    microanàlisi per dispersió d’energia de raigs X (EDS) per determinar la composició química de

    les fibres i la seva morfologia amb mostres prèviament preparades per la doctoranda. A més, la

    doctoranda va participar en el mostreig, extreure el contingut estomacal i va mesurar les fibres

    de vidre, analitzar les dades i redactar l’article, assessorada pels coautors (M. Ramón y M.

    Delgado).

    Chapter IV. Article: “The effects of feeding Karlodinium veneficum (PLY # 103;

    Gymnodinium veneficum Ballantine) to the blue mussel Mytilus edulis”

    publicat a la revista Harmful algae, l’índex d’impacte l’any 2008 fou de 2.688.

    L’experiment es va dur a terme en el Marine Biological Association, Plymouth (UK) sota la

    supervisió del Dr. Pipe. El disseny experimental es va fer conjuntament entre el Dr. Pipe, la

    directora i la doctoranda. L’anàlisi de toxines de Karlodinium veneficum (PLY # 103) el va dur

    a terme el Dr. Place en el seu laboratori de USA. La doctoranda i el Dr. Pipe es van encarregar

    de la recollida dels animals experimentals. La doctoranda va realitzar el muntatge i manteniment

    dels aquaris experimentals, la recollida de mostres, l’anàlisi de dades i la seva interpretació

    (excepte el de toxines), així com de la redacció de l’artícle, assesorada pels coautors (A. R.

    Place, M. Ramón, M. Jutson and R. K. Pipe).

    Chapter V. Article: “Pathology and immune response of the blue mussel (Mytilus

    edulis L.) after an exposure to the harmful dinoflagellate Prorocentrum minimum”

    publicat a la revista Harmful algae, l’índex d’impacte l’any 2008 fou de 2.688. El

    treball experimental es va dissenyar pel Dr. Wikfors, la doctoranda i la tutora de tesi, i es va

    realitzar en el NOAA Milford Laboratory (USA), sota la supervisió del Dr. Wikfors, durant una

    estada de la doctoranda en aquest laboratori. La Dra. Hégaret va ensenyar a la doctoranda l’ús

    del citòmetre de flux i va col·laborar amb ella en l’anàlisi de les mostres. La doctoranda va

    avaluar els canvis histopatològics dels teixits dels musclos sota la supervisió de la Dra. Sunila.

    La doctoranda es va encarregar del manteniment dels aquaris, de dur a terme l’experiment, de

    l’elaboració de les preparacions histològiques, de l’anàlisi de les dades i de la seva interpretació,

    així com de la redacció de l’article, assessorada pels coautors (I. Sunila, H. Hegaret, M. Ramón,

    G. H. Wikfors).

    22

  • Report of the director

    23

    Chapter VI. Article: “Experimental exposure of the blue mussel (Mytilus edulis, L.)

    to the toxic dinoflagellate Alexandrium fundyense: histopathology, immune responses, and

    recovery”

    publicat a la revista Harmful algae, l’índex d’impacte l’any 2008 fou de 2.688. El

    treball experimental es va dissenyar pel Dr. Wikfors, la doctoranda i la tutora de tesi, i es va

    realitzar en el NOAA Milford Laboratory (USA), sota la supervisió del Dr. Wikfors, durant una

    estada de la doctoranda. La Dra. Hégaret va ensenyar a la doctoranda l’ús del citòmetre de flux i

    va col·laborar amb ella en l’anàlisi de les mostres. La doctoranda va avaluar els canvis

    histopatològics dels teixits dels musclos sota la supervisió de la Dra. Sunila. La doctoranda es

    va encarregar del manteniment dels aquaris, de dur a terme l’experiment, de la preparació

    posterior de les mostres d’histologia, de l’anàlisi de les dades i de la seva interpretació, així com

    de la redacció de l’article, assessorada pels coautors (I. Sunila, H. Hegaret, M. Ramón, G. H.

    Wikfors).

  • RESULTS

    24

  • Chapter I

    Feeding behavior of the mussel Mytilus galloprovincialis (L.) in a

    Mediterranean estuary: a field study

    Submitted

    25

  • 26

    Feeding behavior of the mussel Mytilus galloprovincialis (L.) in a Mediterranean estuary: a

    field study.

    Eve Galimany1, 3, 5, *, Montserrat Ramón2, 3, 5, Irrintzi Ibarrola4

    1 IRTA, Crta. Poble Nou s/n St. Carles de la Ràpita 43540, Spain; 2 IEO-Centre Oceanogràfic de les

    Balears, Moll de Ponent s/n, Palma 07015, Spain; 3 ICM-CSIC, Psg. Marítim de la Barceloneta 37-49,

    Barcelona 08003, Spain; 4 Dpto. Genética, Antropología Física y Fisiología Animal, Facultad de Ciencia

    y Tecnología, Universidad del País Vasco/Euskal Herriko Unibertsitatea, Bilbao 48080, Spain; 5 Xarxa de

    Referència de Recerca i Desenvolupament en Aqüicultura de la Generalitat de Catalunya, XRAq, Spain

    *Corresponding author: Eve Galimany

    Ph: (0034) 977 745 427; Fax: (0034) 977 744 138; e-mail: [email protected]

    Key words: feeding behavior; shellfish; physiology; field experiment; Alfacs Bay; seston.

    Abstract

    The feeding behavior of the mussel

    Mytilus galloprovincialis was studied under

    field conditions on top of a mussel raft in

    Alfacs Bay, N.W. Mediterranean Sea. The

    experiments were performed in November

    2006 and February, April and July 2007

    using a flow-through filter feeding device.

    Total particulate matter (TPM), particulate

    organic matter (POM) and particulate

    inorganic matter (PIM) were calculated for

    the bay water, and the feces and pseudofeces

    of the mussels, to obtain different feeding-

    physiological parameters through the

    biodeposition method, such as the clearance

    rate (CR), organic ingestion rate (OIR),

    absorption rate (AR) and absorption

    efficiency (AE). The device employed for the

    experiments was validated before the data

    was analyzed. The results showed that short-

    term variations in food quantity and quality

    were similar to the long-term variations;

    nevertheless, we found a high organic

    content of the bay water throughout, with f

    (POM/TPM) values ranging from 0.51 to

    0.72, which is comparable to other areas with

    high mussel aquaculture production.

    Physiological parameters characterizing both

    food acquisition and absorption were found

    to vary greatly in the short-term (days).

    Nevertheless, we found high CR and AE

    values throughout the study. As a response to

    the variable environment of Alfacs Bay,

    mussels reduced their clearance rate when

    seston concentrations were high instead of

    increasing their pseudofeces production.

    Absorption efficiency was positively related

    to the organic content of the ingested

    particles (i), which has been shown

    previously to be a main determinant of

    absorption efficiency. We could not

    determine a seasonal pattern but there was a

  • clear tendency for mussels to decrease their

    CR, OIR and AR during July. It is possible

    that this feeding restriction was due to the

    high temperature of the bay water in that

    month, which reached 25ºC.

    Our results show that the quality of

    seston in Alfacs Bay is good for bivalve

    farming. The physiological parameters

    measured were high, except in July, when

    mussels were negatively influenced by the

    high water temperature. This type of study is

    very useful for the management of bivalve

    aquaculture sites because data is obtained

    from a field study.

    1. Introduction

    Mussels (Mytilus galloprovincialis)

    are cultured on the Spanish Mediterranean

    coast in the two Ebro Delta bays, Fangar and

    Alfacs, with an annual production of 3000 t

    per year (Ramón et al., 2005a). They are

    cultured in suspension from a total of 166

    fixed rafts divided between the bays of

    Alfacs (90 farms) and Fangar (76 farms). The

    rafts are rectangular wooden frames

    measuring 200 m x 15 m, from which 2 to 3

    m long mussel ropes are suspended. There is

    a seasonal pattern for the growth of mussels

    in Fangar Bay, with higher rates in spring

    and the lowest rates in winter and August.

    The growth and mortality of mussels cultured

    in the two delta bays are strongly affected by

    the high seawater temperatures reached

    during July and August, which cause a

    detention in growth and high mortalities of

    adults and juveniles (Ramón et al., 2005a).

    Mussels (Mytilus spp.) are filter

    feeding bivalves that feed on a suspension of

    water particles. Areas that have bivalve

    cultures are characterized by high primary

    production that sustains their grazing

    pressure. Facing the Mediterranean Sea,

    Thau Lagoon (S.E. France) has high primary

    production (Gasc, 1997), with particulate

    organic matter (POM) values ranging from

    0.1 to 1.7 mg L-1 (Gangnery et al., 2004).

    The study site, Alfacs Bay (N.E. Spain), has

    a chlorophyll a level that is one order of

    magnitude above the surrounding

    Mediterranean Sea (Delgado, 1987), and

    POM values ranging from 1 to 3.4 mg L-1

    (Ramón et al., 2005b). Moreover, the POM

    concentration in Alfacs Bay is even higher

    than in the Rías Gallegas (N.W. Spain), the

    largest mussel producer in Europe (Pérez

    Camacho et al., 1991; Keldany, 2002). As an

    example, POM concentrations found in Ría

    de Arousa, the most important of the

    Galician Rias for mussel production, range

    from 0.28 to 1.08 mg L-1 (Navarro et al.,

    1991; Babarro et al., 2003). In addition to the

    POM, food quality expressed as POM/TPM

    can have a significant influence on the

    growth of bivalves, which gives an idea of

    the relative importance of the organic matter

    in the environment. There is a high

    percentage of organic content in the water of

    both Alfacs Bay and Ría de Arousa, and

    values can reach up to 50-60% (Ramón et al.,

    2005b; Babarro et al., 2003).

    The feeding behavior of mussels has

    been studied by means of different

    27

  • approaches. Foster-Smith (1975) calculated

    filtration rates from the particle removal rates

    from a fixed volume of suspension. For a

    better approach to natural conditions, other

    authors estimated clearance rates by

    monitoring the removal of suspended

    particles as water passed through mussels;

    therefore, the filtration rates could be

    calculated as the clearing of water particles

    from the environmental water (Riisgård,

    1977; Bayne and Widdows, 1978; Widdows

    et al., 1979). The need to simulate real

    conditions in order to better understand the

    feeding behavior in the field has led

    researchers to design new devices. Smith and

    Wikfors (1998) developed an automated

    rearing chamber system for studies of

    shellfish feeding behavior, and Babarro et al.

    (2000) used a portable box raft experimental

    chamber to conduct experiments in situ on

    top of mussel rafts. Different portable filter

    feeding devices have been designed in order

    to better understand the feeding

    physiological parameters of mussels

    (Filgueira et al., 2006; Grizzle et al., 2006).

    Nowadays, there are different methods for

    measuring filtration rates in suspension

    feeding bivalves, each of which should be

    used according to the experimental

    conditions and taking into account their

    advantages and disadvantages (reviewed by

    Riisgård, 2001).

    The aim of the present study is to use

    the biodeposition method to determine the

    main physiological parameters related to

    mussel feeding behavior in a Mediterranean

    estuary where bivalve aquaculture takes

    place (Iglesias et al., 1998). The experiments

    were carried out on top of a mussel raft at

    four different periods of the year for a more

    realistic approach. These results will be

    useful for evaluating the carrying capacity of

    Alfacs Bay for bivalve growth, and for

    production management using simulation

    models.

    2. Materials and methods

    2.1 Experimental design and animals

    The filter feeding experiments were

    performed from November 2006 to July

    2007. Four filter feeding experiments, lasting

    2 hours each, were carried out per sampling

    period (i.e., November 2006, and February,

    April and July 2007), corresponding to 2

    consecutive days in 2 consecutive weeks in

    each period, except for July, in which 3

    experiments were carried out.

    Mussels, Mytilus galloprovincialis,

    were collected from a mussel aquaculture

    farming site in Alfacs Bay (Ebro Delta) the

    day before each experiment. Twenty mussels

    per experiment were collected, and epiphytes

    and other encrusting organisms were

    removed from the shells. A little plastic hook

    and loop fastener was glued to one of the two

    shells of each individual. When the glue

    dried, mussels were hung back on the raft

    where the experiments were performed the

    following day. Acclimation was not

    necessary as the mussels were always

    submerged in the bay water. Fake mussels

    were made by collecting four of the twenty

    28

  • 29

    fresh mussels, taking out the flesh and gluing

    the shells back together to act as controls.

    Two portable filter feeding flow-

    through devices were designed to simulate in

    vivo conditions of mussel feeding (Fig. 1). 2.2 Flow-through devices

    Mussel raft surface

    Under water

    contro

    lco

    ntro

    l

    aeration

    cont

    rol

    contro

    l

    aeration

    1

    23

    4

    5

    6

    7

    4

    23

    5

    6

    7

    Fig. 1: Diagram of the two portable filter feeding devices. 1: underwater pump; 2: common PVC tanks; 3:

    common tank extra flow exit tube; 4: aeration; 5: rubber tubes; 6: aquarium extra flow exit tubes; 7:

    manual valves.

    One portable filter feeding device

    consisted in a common PVC tank (length 560

    mm, width 300 mm, height 150 mm) that

    received bay water from an underwater pump

    hung from the mussel raft poles. Laterally,

    the tank was provided with an extra flow exit

    tube. Aeration was provided to mix the bay

    water in the common tank. Ten rubber tubes

    emerged from the lower part of the tank.

    Each tube was connected to an individual

    PVC aquarium. The aquaria measured 45 x

    180 x 60 mm (length x width x height), and

    each aquarium contained a single mussel.

    Two of the aquaria contained one fake

    mussel each, which were used as controls.

    The mussels were positioned near the flow

    exit tube of the aquaria and attached to the

    bottom with a piece of plastic hook and loop

    fastener. The flow of bay water from the

    common tank to each aquarium was

    regulated through a manual valve and

    maintained at 12 L h-1. This flux was

    determined by previous laboratory

    experiments to validate the optimal geometry

    chamber using the flow-through chamber

    method, following Riisgård (1977). A flow

  • of 12 L h-1 showed a homogeneous

    distribution of particles between aquaria and

    no water recirculation occurred in any of the

    aquaria. These filter feeding devices were

    deployed in the same area that the mussel

    cultures are located, i.e., on a mussel raft.

    2.3 Characteristics of the bay water

    Bay water (1 L) from the tanks with

    fake mussels (acting as controls) was

    collected every 15 minutes, filtered

    separately through washed Whatman GF/C

    filters (25 mm diameter circles) and rinsed

    with ammonium formate. In the laboratory,

    filters with water samples were dried at 60ºC

    for 48 h and weighed to obtain the dry

    weight, which accounted for the total

    particulate matter (TPM). Afterwards, filters

    were ashed at 450ºC for 4 h before the final

    weighing to obtain the particulate inorganic

    matter (PIM). The particulate organic matter

    (POM) was calculated as the weight loss

    between TPM and PIM. The organic content

    of the bay water (f) was calculated as the

    fraction between POM and TPM.

    2.4 Physiological feeding parameters

    The individual chambers of the

    feeding-devices were cleaned before the

    collection of biodeposits for the experiments.

    Then, the feces and pseudofeces of each

    mussel were collected separately with a

    pipette as soon as they were produced and

    accumulated in each individual filter. After

    approximately 2 h the mussels had produced

    a large amount of biodeposits ending the

    collection period of the experiment.

    Filters with samples of feces and

    pseudofeces produced by each individual

    mussel (n= 16) were processed for organic

    and inorganic matter as indicated for the

    water samples, in order to compute the total,

    organic and inorganic egestion and rejection

    rates. The physiological components

    Table 1: Physiological components of absorptive balance for mussels in Alfacs Bay.

    Parameter Acronym Units Calculation

    Clearance rate CR L h-1 (mg inorganic matter from both feces and pseudofeces per unit of time) / (mg inorganic matter available in bay water L-1)

    Filtration rate FR mg h-1 CR x mg total matter available in L-1 bay water

    Rejection rate RR % [(mg total matter within pseudofeces deposited h-1) / FR] x

    100 Organic ingestion rate OIR mg h

    -1 (CR x mg organic matter available in L-1 bay water) – (mg organic matter within pseudofeces deposited h-1)

    Absorption rate AR mg h-1 OIR - (mg organic matter within feces deposited h-1)

    Absorption efficiency AE fraction AR / OIR

    Selection efficiency SE fraction

    1 – [(organic fraction within pseudofeces) / (organic fraction within total particles available in bay water)]

    30

  • of the absorptive balance (Table 1) were then

    calculated according to the biodeposition

    method (Iglesias et al, 1998).

    The ingestion rates of total matter

    (TIR: mg/h) and organic matter (OIR: mg/h)

    were obtained as the difference between the

    filtration and rejection rates of either the total

    or organic matter. In order to quantify the

    preingestive rejection of food through

    pseudofeces production, we expressed

    rejection of food as a percentage of filtered

    matter. Finally, the organic content of

    ingested matter was calculated as OIR/TIR.

    It was necessary to calculate the gut

    transit time (GTT) in order to compare the

    seston ingested with the corresponding

    biodeposits of the mussels, as the food

    digestion process takes some time.

    Therefore, the GTT was calculated before

    each experiment using a method adapted

    from Hawkins et al. (1996). Five mussels

    were placed individually in a beaker in a

    mixture of bay water and Tetraselmis suecica

    monoculture. The time that elapsed between

    the beginning of the exposure and the

    deposition of green colored feces was

    considered to be the GTT (min).

    All parameters were standardized to

    1 g of dried mussel flesh using the following

    equation:

    Ys = Ye x (1/We) b

    where Ys is the standardized physiological

    rate, Ye is the experimentally determined rate

    and We is the measured dry body mass of

    each mussel. We used a b value of 0.67,

    which has been used in previous mussel

    feeding studies (Bayne et al., 1989, 1993;

    Jones et al., 1992; Hawkins et al., 1997).

    2.5 Statistical analyses

    Data were checked for normality and

    variance homogeneity. A non-parametric test

    (Kolmogorov-Smirnov) was used to compare

    the results obtained for the TPM, POM, PIM

    and the organic content of the bay water (f)

    with the two filter feeding devices. The

    results for TPM, POM, PIM and the

    percentage of organic content of the bay

    water, gut transit time (GTT) and the in situ

    filter feeding parameters were compared

    between sampling periods using the Kruskal-

    Wallis one-way Analysis of Variance. The

    statistical software used was Statgraphics

    Plus (Manugistics, Inc., Rockville, MD,

    USA). Correlations were established between

    f and TPM, CR and TPM, and AE and i

    (organic content of the ingested material)

    using non-linear regression models and the

    statistical software SPSS Statistic 17.0 (SPSS

    Inc, Chicago).

    3. Results

    3.1 Bay water

    The mean values of the bay water

    temperature (ºC) for November, February,

    April and July during the experiments were

    18.1 ± 0.1, 10.56 ± 0.1, 16.7 ± 0.1 and 25.9 ±

    0.1 respectively.

    TPM, POM, PIM and the organic

    content of seston (f) of the bay water showed

    no significant differences between the two

    flow-through devices when all the samplings

    were analyzed together (p>0.05). Mean

    31

  • 32

    values for the different parameters for the

    two devices together throughout the study

    period were TPM: 1.67 ± 0.09; POM: 0.99 ±

    0.04; PIM: 0.68 ± 0.06, expressed as mg L-1;

    and f: 0.62 ± 0.01. Nevertheless, when the

    two devices were compared for each

    sampling day and period of the study, TPM,

    POM and PIM showed significant

    differences for April and day 9 (the first

    sampling day of April) (p0.05). Therefore, day 9

    was not taken into account in the study of

    mussel filter feeding parameters, as the

    devices were not comparable on that day for

    unknown reasons.

    Sampling day TPM (mg L-1) POM (mg L-1) PIM (mg L-1) f (POM/TPM)

    November 2006

    1 1.63 ± 0.13 1.03 ± 0.05 0.60 ± 0.07 0.65 ± 0.02

    2 1.41 ± 0.13 0.88 ± 0.05 0.55 ± 0.08 0.63 ± 0.02

    3 1.30 ± 0.13 0.79 ± 0.05 0.51 ± 0.08 0.62 ± 0.02

    4 2.09 ± 0.13 1.14 ± 0.05 0.88 ± 0.08 0.54 ± 0.02

    February 2007

    5 2.17 ± 0.16 1.06 ± 0.05 1.11 ± 0.08 0.51 ± 0.02

    6 1.51 ± 0.16 0.85 ± 0.06 0.42 ± 0.08 0.66 ± 0.02

    7 1.21 ± 0.17 0.71 ± 0.05 0.50 ± 0.09 0.62 ± 0.02

    8 1.02 ± 0.17 0.67 ± 0.05 0.35 ± 0.08 0.67 ± 0.02

    April 2007

    10 1.72 ± 0.10 0.99 ± 0.05 0.73 ± 0.08 0.58 ± 0.02

    11 2.00 ± 0.18 1.27 ± 0.09 0.74 ± 0.14 0.64 ± 0.04

    12 1.71 ± 0.15 1.12 ± 0.08 0.60 ± 0.12 0.66 ± 0.03

    July 2007

    13 2.30 ± 0.14 1.30 ± 0.06 0.91 ± 0.09 0.58 ± 0.03

    14 2.16 ± 0.17 1.36 ± 0.07 0.80 ± 0.10 0.61 ± 0.03

    15 1.59 ± 0.18 1.15 ± 0.07 0.44 ± 0.11 0.72 ± 0.03

    Table 2: Mean values (±SE) of total particulate matter (TPM), particulate organic matter (POM),

    particulate inorganic matter (PIM) and quality of seston (f) of the bay water for each sampling day.

    The mean values for TPM, POM,

    PIM and f are shown in Table 2. Mean values

    of TPM ranged from approximately 1 to 2.3

    mg L-1. February showed significantly lower

    values of TPM, whereas April and July

    showed the highest seston concentration. The

    organic content of particles (f) ranged from

    0.51 to 0.72 with no significant differences

    between sampling periods (p>0.05). There

    was a negative relationship between organic

    content (f) and total particulate matter (TPM)

    in the bay water throughout the entire study

  • period, (f = 0.827 (±0.021) e -0.183 (±0.014); r² =

    0,466) (p0.05).

    Figure 3 shows the relationship

    between GTT and water temperature. The

    linear regression indicates that temperature

    explains 30% of the GTT variation (p0.05). Physiological parameters that

    characterize both food acquisition and

    absorption were found to change greatly in

    the short-term (days). For instance, whereas

    the total annual variation of the clearance

    rate ranged from a minimum of 1.06 L h-1

    (July) to a maximum of 4.83 L h-1

    (February), on two different days of

    November the CR ranged from 1.51 to 4.34

    L h-1. There seems to be a downward trend of

    Sampling period GTT (min)

    November 06 71.82 ± 1.35

    February 07 82.09 ± 1.30

    April 07 67.29 ± 1.52

    July 07 61.70 ± 1.47

    33

  • Table 4: Mean values (±SE) of clearance rate (CR), filtration rate (FR), rejection rate (RR), selection efficiency (SE), organic ingestion rate (OIR), absorption rate (AR)

    and absorption efficiency (AE) for each sampling day.

    Sampling day CR (L h

    -1) FR (mg h-1) RR (%) SE OIR (mg h-1) AR (mg h-1) AE

    November 2006

    1 3.02 ± 0.38 6.21 ± 0.62 2.92 ± 2.36 0.22 ±0.18 3.01 ± 0.35 2.40 ± 0.28 0.74 ± 0.02

    2 4.34 ± 0.39 6.11 ± 0.64 4.51 ± 2.44 0.72 ± 0.10 3.58 ± 0.36 2.65 ± 0.29 0.73 ± 0.02

    3 3.21 ± 0.39 4.12 ± 0.64 1.38 ± 2.36 0.88 ± 0.10 2.47 ± 0.36 1.70 ±0.29 0.67 ± 0.02

    4 1.51 ± 0.85 3.07 ± 1.39 9.70 ± 5.27 0.75 ± 0.23 1.58 ± 0.78 1.03 ± 0.63 0.62 ± 0.05

    February 2007

    5 3.17 ± 0.46 7.24 ± 0.76 4.23 ± 2.89 0.77 ± 0.13 3.33 ± 0.43 2.18 ± 0.34 0.61 ± 0.03

    6 3.17 ± 0.38 4.56 ± 0.62 1.27 ± 2.36 0.83 ± 0.10 2.75 ± 0.35 2.08 ± 0.28 0.73 ± 0.02

    7 2.31 ± 0.37 2.68 ± 0.60 0 1 2.68 ± 0.60 1.09 ± 0.27 0.65 ± 0.02

    8 4.83 ± 0.41 4.99 ± 0.67 0.22 ± 2.53 0.57 ± 0.11 3.11 ± 0.38 2.50 ± 0.30 0.80 ± 0.02

    April 2007

    10 2.16 ± 0.37 2.33 ± 0.60 0 1 2.33 ± 0.60 1.48 ± 0.27 0.62 ± 0.02

    11 2.29 ± 0.50 4.53 ± 0.80 7.71 ± 3.04 0.60 ± 0.13 2.65 ± 0.45 1.71 ± 0.36 0.64 ± 0.03

    12 1.97 ± 0.41 3.26 ± 0.67 2.73 ± 2.53 0.73 ± 0.11 2.06 ± 0.38 1.38 ± 0.30 0.65 ± 0.02

    July 2007

    13 2.16 ± 0.38 4.91 ± 0.62 4.33 ± 2.35 0.65 ± 0.10 2.62 ± 0.35 1.57 ± 0.28 0.60 ± 0.02

    14 1.06 ± 0.46 2.32 ± 0.76 11.61 ± 2.89 0.49 ± 0.13 1.22 ± 0.43 0.75 ± 0.34 0.60 ± 0.03

    15 1.06 ± 0.42 1.60 ± 0.70 13.23 ± 2.31 0.60 ± 0.11 1.06 ± 0.39 0.73 ± 0.31 0.67 ± 0.02

    34

  • the CR, OIR, AR and AE values from

    November and April to July (Fig. 4). Table 5

    shows significant associations between the

    different sampling periods for CR, OIR, AR

    and AE. The maximum values for CR were

    obtained in November and February, and the

    mean values for these sampling periods were

    not significantly different (p>0.05). In

    contrast, April and July showed significantly

    lower values compared to November and

    February (p

  • Fig. 5: Clearance rate (CR) related to total

    particulate matter (TPM) during the different

    periods of the study (Nov: November 2006; Feb:

    February 2007; Apr: Aril 2007; Jul: July 2007).

    Exponential relationship shown in the graph.

    In agreement with previous authors,

    the absorption efficiency was found to be a

    positive function of the organic content of

    ingested material (i), described by an

    asymptotic exponential. Although variability

    of both AE and food organic content is

    relatively small in these experiments, we

    have fitted our results to asymptotic

    exponential curves (Fig. 6) and the following

    models were obtained for the different

    sampling periods:

    November:

    AE = 0.950 (±0.143) (1-e (-2.180 (±0.706)xi));

    r2 = 0.334, n= 46

    February:

    AE = 0.950 (±0.116) (1-e (-2.432 (±0.666)xi));

    r2 = 0.586, n= 38

    April:

    AE = 0.950 (±0.662) (1-e (-1.751 (±2.339)xi));

    r2= 0.237, n= 22

    July:

    AE = 0.825 (±0.119) (1-e (-2.196 (±0.711)xi));

    r2= 0.228, n= 38

    Mussels had a lower capacity to

    absorb available food during July, which is

    shown by the fact that the lowest asymptotic

    AE value was recorded for this period.

    0123456

    0,5 1,0 1,5 2,0 2,5

    CR

    (L h

    -1)

    TPM (mg L-1)

    Nov Feb Apr Jul

    0,3

    0,5

    0,7

    0,9

    1,1

    0,4 0,6 0,8 1,0

    AE

    i

    Nov Feb Apr Jul

    Fig. 6: Absorption efficiency (AE) related to the

    organic content of the ingested material (i) for

    each individual mussel.

    Discussion

    The methodology used for bivalve

    feeding experiments is a key issue for

    obtaining valuable data. Petersen et al.

    (2004) found significantly lower clearance

    rates measured with the biodeposition

    method compared to those measured with the

    flow-through and indirect methods.

    However, these last two methods considered

    shorter periods of time, when mussels were

    actively feeding, whereas the biodeposition

    method includes time periods of potential

    inactivity (Bougrier et al., 1998; Iglesias et

    al., 1998). Therefore, the biodeposition

    method is recommended for studying

    dynamic feeding responses in natural

    environments, as it integrates over time and

    enables several feeding parameters to be

    calculated (Pascoe et al., 2009), as we have

    done in this study.

    36

  • November February April July CR OIR AR AE

    Table 5: Summary of the Kruskal-Wallis one way Analysis of Variance among physiological

    determinations over time. CR: Clearance Rate; OIR: Organic Ingestion Rate; AR: Absorption Rate; AE:

    Absorption Efficiency. Straight lines indicate no significant differences (p>0.05).

    Food characteristics (quantity and

    quality) during the experiments were found

    to be rather high during the study and similar

    to the values previously obtained in the area

    (Ramón et al., 2005b). Short-term variation

    (days) is distinctive of Alfacs Bay, and can

    be as large as the variability found for the

    different periods; it is mainly influenced by

    the wind regime and the shallow depth (5 m

    maximum). The negative relationship found

    for the bay water between organic content (f)

    and total particulate matter (TPM) is

    probably due to resuspension phenomena, as

    reported in other estuarine ecosystems, which

    would add silt and inorganic material to the

    water column (Hawkins et al., 1996; Velasco

    and Navarro, 2005). The seston

    characteristics from Alfacs Bay, on average,

    are similar to other aquaculture sites, but the

    minimum values of TPM and POM were

    higher than in Ría de Arosa, the largest

    mussel producer in Europe (Navarro et al.,

    1991; Babarro et al., 2000, 2003). Our results

    indicate that the seston (f) is of good quality

    with an average value of 0.62. In addition, in

    Alfacs Bay, f is also higher than reported for

    the Galician Rías (Babarro et al., 2000,

    2003). Therefore, it is evident that Alfacs

    Bay is an ideal area for mussel farming in

    terms of food quality, and that the organic

    content of the seston is unlikely to be a

    limiting factor for this industry with the

    current mussel production.

    The physiological parameters related

    to filter-feeding activities were found to

    change greatly in the short-term. However,

    the wide range of values found in the long-

    term for CR and OIR throughout the study

    has also been reported in other estuarine

    aquaculture sites (Babarro et al., 2000). As a

    response to the variable environment of

    Alfacs Bay in which the seston concentration

    is not constant, bivalves possess regulatory

    mechanisms. Foster-Smith (1975) proposed

    two mechanisms, which can occur together

    or separately: 1) a clearance rate reduction,

    and 2) an increase in pseudofeces production.

    Since the food rejection rate was low and

    constant throughout our study, the negative

    relationship between the clearance rate and

    the particle concentration becomes the main

    determinant for the food ingestion rate, as

    reported elsewhere (Widdows et al., 1979;

    Babarro et al., 2000). However, other authors

    37

  • have obtained the opposite results, in which

    mussels increased the production of

    pseudofeces instead of regulating the

    clearance rate (Foster-Smith, 1975; Navarro

    et al., 2003). In these cases, the quality of

    food was low, in contrast to what we found

    in Alfacs Bay, and it seems that bivalves

    increase the rejection rates when they feed on

    seston with low organic content (Navarro et

    al., 1992; Urrutia et al., 1996; Navarro et al.,

    2003). Nevertheless, it is not clear whether

    the different strategies used to regulate the

    feeding behavior in relation to the seston

    concentration variability are species-specific

    or habitat-specific (Bacon et al., 1998).

    Absorption efficiency was positively related

    to the organic content of the ingested

    particles (i), which has been previously

    shown to be a main determinant of

    absorption efficiency (Hawkins et al., 1998).

    The relationship between AE and the quality

    of seston has also been documented for

    mussels in Marennes-Oléron Bay and in the

    Galician Rías (Navarro et al., 1991; Hawkins

    et al., 1996; Babarro et al., 2000), where AE

    is clearly dependent on food quality.

    We observed a tendency for the

    physiological parameters characterizing both

    food acquisition and absorption to decrease

    during April and July. CR in April was lower

    than in the previous two sampling periods

    but, as OIR was not significantly different

    from that of November and February, this

    suggests an ingestion regulation phenomenon

    due to a high seston concentration in the bay

    water. July was the period with the lowest

    CR, OIR and AR. Although similar values of

    TPM were obtained during all the sampling

    periods, the CR of mussels decreased in July.

    Similarly, a decrease in the asymptotic AE

    value recorded for July compared with the

    remaining sampling periods (0.825 vs. 0.950)

    suggests that the mussels have a relatively

    lower capacity to absorb the available food

    during July. Low ingestion rates would be

    coupled with high GTT and high AE as an

    acclimation response to high seston quantity

    (Bayne et al., 1987). However, and although

    we found no significant differences between

    GTT during the study, these values were

    lowest in July and ingestion rates were also

    the lowest. It is known that ectothermic

    animals, such as bivalves, depend on the

    water temperature for many of their

    physiological processes. The inverse

    relationship found between GTT and

    temperature might indicate that this

    environmental parameter is responsible for

    restricting the functional capacity of mussels

    during July. The thermodependance of the

    CR of mussels has been previously reported

    (Denis et al., 1999; Babarro et al., 2000). The

    temperature threshold at our experiment site,

    despite a difference of 16ºC during the study,

    is in the tolerance range of M.

    galloprovincialis (Anestis et al., 2007).

    Nevertheless, Hofmann and Somero (1996)

    reported that the threshold induction

    temperature for stress proteins for M.

    galloprovincialis is 25ºC, and we recorded a

    water temperature of 25.98±0.14 during July.

    Therefore, temperature is probably the factor

    38

  • that most influences the decrease in the

    feeding parameter values for mussels at the

    study site during July.

    As a summary, we can conclude that

    Alfacs Bay is an ideal site for developing

    mussel aquaculture due to its high water

    quality, which seems to be stable throughout

    the year. Similarly, filtration activity and

    absorption efficiency were also high

    throughout the studied periods, and depend

    on TPM and the organic content of the

    ingested particles respectively. The lower

    feeding capacity observed during July seems

    to be influenced by high water temperatures,

    which would cause physiological stress on

    the mussels. This study contributes to a better

    understanding of in situ feeding behavior of

    M. galloprovincialis in Alfacs Bay, which is

    useful for further studies on the carrying

    capacity of the ecosystem.

    Acknowledgements

    This study has been supported by a

    grant from the Instituto Nacional de

    Investigación y Tecnología Agraria y

    Alimentaria (INIA) to the first author and

    partially financed by the RTA04-023 INIA

    research project. We would like to thank

    Xavier Mas, Xavier Leal and David Pina

    from ICM-CSIC for their advices and help

    during the construction of the flow-through

    devices. We would also like to acknowledge

    J. M. Reverté, Vanesa Castán and Esther

    Dámaso for their help during the field work.

    References

    Anestis, A., Lazou, A., Pörtner, H.O.,

    Michaelidis, B., 2007. Behavioral,

    metabolic, and molecular stress

    responses of marine bivalve Mytilus

    galloprovincialis during long-term

    acclimation at increasing ambient

    temperature. Am. J. Physiol. Regul.

    Integr. Comp. Physiol. 293, R911–

    R921.

    Babarro, J.M.F., Fernández-Reiriz, M.J.,

    Labarta, U., 2000. Feeding behavior

    of seed mussel Mytilus

    galloprovincialis: environmental

    parameters and seed origin. J.

    Shellfish Res. 19, 195-201.

    Babarro, J.M.F., Fernández-Reiriz, M.J.,

    Labarta, U., 2003. In situ absorption

    efficiency processes for the cultured

    mussel Mytilus galloprovincialis in

    Ría de Arousa (north-west Spain). J.

    Mar. Biol. Ass. U. K. 83, 1059-1064.

    Bacon, G.S., MacDonald, B.A., Ward, J.E.,

    1998. Physiological responses of

    infaunal (Mya arenaria) and

    epifaunal (Placopecten

    magellanicus) bivalves to variations

    in the concentration and quality of

    suspended particles I. Feeding

    activity and selection. J. Exp. Mar.

    Biol. Ecol. 219, 105-125.

    Bayne, B.L., Hawkins, A.J.S., Navarro, E.,

    1987. Feeding and digestion by the

    mussel Mytilus edulis L. (Bivalvia:

    Mollusca) in mixtures of silt and

    39

  • algal cells at low concentration J.

    Exp. Mar. Biol. Ecol. 111, 1-22.

    Bayne, B.L., Hawkins, A.J.S., Navarro, E.,

    Iglesias, J.I.P., 1989. Effects of

    seston concetration on feeding,

    digestion and growth in the mussel,

    Mytilus edulis. Mar. Ecol. Prog. Ser.

    55, 47-54.

    Bayne, B.L., Iglesias, J.I.P., Hawkins, A.J.S.,

    Navarro, E., Heral, M., Deslous

    Paoli, J.M., 1993. Feeding behaviour

    of the mussel, Mytilus edulis:

    responses to variations in quantity

    and organic content of the seston. J.

    Mar. Biol. Ass. U. K. 73, 813-829.

    Bayne, B.L., Widdows, J., 1978. The

    physiological ecology of two

    populations of Mytilus edulis L.

    Oecologia 37, 137-162.

    Bougrier, S., Collet, B., Geairon, P., Geffard,

    O., Héral, M., Deslous Paoli, J.M.,

    1998. Respiratory time activity of the

    Japanese oyster Crassostrea gigas

    (Thunberg). J. Exp. Mar. Biol. Ecol.

    219, 205-216.

    Delgado, M., 1987. Fitoplancton de las

    bahías del delta del Ebro. Inv. Pesq.

    51, 517-548.

    Denis, L., Alliot, E., Grzebyk, D., 1999.

    Clearance rate responses of

    Mediterranean mussels, Mytilus

    galloprovincialis, to variations in the

    flow, water temperature, food quality

    and quantity. Aquat. Living Resour.

    12, 279-288.

    Filgueira, R., Labarta, U., Fernández-Reiriz,

    M.J., 2006. Flow-through chamber

    method for clearance rate

    measurements in bivalves: design

    and validation of individual

    chambers and mesocosm. Limnol.

    Oceanogr.: Methods 4, 284-292.

    Foster-Smith, R.L., 1975. The effect of

    concentration of suspension on the

    filtration rates and pseudofaecal

    production for Mytilus edulis L.,

    Cerastoderma edule (L.) and

    Venerupis pullastra (Montagu). J.

    Exp. Mar. Biol. Ecol. 17, 1-22.

    Gangnery, A., Bacher, C., Buestel, D., 2004.

    Application of a population

    dynamics model to the

    Mediterranean mussel, Mytilus

    galloprovincialis, reared in Thau

    Lagoon (France). Aquaculture 229,

    289-313.

    Gasc, A., 1997. Study of regenerated primary

    production in a mollusc culture

    ecosystem: the Thau lagoon,

    Montpellier (France), 216 pp.

    Grizzle, R.E., Greene, J.K., Luckenbach,

    M.W., Coen, L.D., 2006. A new in

    situ method for measuring seston

    uptake by suspension-feeding

    bivalve molluscs. J. Shellfish Res.

    25, 643-649.

    Hawkins, A.J.S., Bayne, B.L., Bougrier, S.,

    Héral, M